the effects of psychosocial deprivation on human ...historically in terms of the consequences of...
TRANSCRIPT
1
ED 034 574
AUTHORTITLE
INSTITUTION
SPANS AGENCY
PUB DATENOTE
EDPS PRICEDESCRIPTORS
IDENTIFIERS
ABSTRACT
DOCUMENT RESUME
PS 002 509
Caldwell, Bettye M.The Effects of Psychosocial Deprivation on HumanDevelopment in Infancy.Center for Early Development and Education, LittleRock, Ark.National Inst. of Child Health and HumanDevelopment, Bethesda, Md.r.68135p.
EDPS Price MF-$0.25 HC-$1.85*Behavioral Science Research, Child Development,Cognitive Development, *Cultural Disadvantagement,Environmental Influences, *Human Development,Infants, Parent Influence, Preschool Children,*Research Needs, *Research Problems, ResearchPeviews (Publications), Social InfluencesPsychosocial Deprivation
The concept of deprivation has become very appealingto specialists of many disciplines as an explanation for deviationsin human development. This is understandable, since the phenomenondoes seem to be a key factor in development, but several technicaland methodological considerations hinder immediate efforts atunderstanding and alleviating it. Ambiguities in definitions of theconcept of deprivation and the necessity to rely on field studyresearch designs constitute major deterrents to the acquisition ofdefinitive data. To determine more clearly the effects ofpsychosocial deprivation on human development there have to be
improved techniques for assessing the psychosocial environment andfor standardizing change-sensitive measures of early development.Also, more attention needs to be given to clarifying the relationshipbetween constitutional factors and susceptibility to deprivation. Thesingle case model deserves more use in attempts to demonstrate theeffects of the operation and removal of psychosocial deprivation.Finally, psychosocial enrichment might have an optimum level whichshould be taken into consideration. (Author/MH)
U. 1, DEPARTMENT or HEAtTH, EDUCATMN t warmOFFICE CF CM 3'TION
THIS Pft,'INE''T f RrC:IVED FROM THE
FEr:) ',!; r 7! OF VIM OR OPI:TIS
1):: \I:!'" OFFICE OF ElMnION
Frain C 2 K.11.10.THE EFFECTS OF PSYCHOSOCIAL DEPRIVATION ON
HUMAN DEVELOPMENT IN INFANCY
Bettye M. Caldwell
University of Arkansas
As an explanation for certain distortions in human development, the concept
of deprivation has been very appealing to scientists from diverse ackgrounds
and with varied areas of application. Whenever any approximation of a
minimal level of necessary input can be: established, the deprivation model
appears to be appropriate as a frame df reference; furthermore, it appears to
be relevant for all living organi ms and to the entire gamut of developmental
influences from biochemical to psychosocial. Also, it seems to carry within
it the promise of remediation, for if one can specify the substance of which
the organism has been deprived, then one should hopefully be able to supply
the missing ingredients in the proper quantities at the correct time.
The natural logic inherent in the concept of deprivation--i.e., that an
insufficiency of a particular substance or experience would lead to untoward
developmental consequences--has stimulated a vast amount of fruitful
research. It is probably accurate to say that in general the sequential
investigations have followed something of an evolutionary pattern in terms
of choice of the dependent variable. The first of these is usually life
itself, with sublethal effects gradually added to the list of possible con-
sequences in an order that progresses from somatic toward psychosocial. Thus
historically in terms of the consequences of nutritional deprivation, early
studies tended to stress the effects of malnutrition on stature or brain
weight, with research into its effects on mental development very current
and that concerned with socio-emotional development more contemplated
VI
2
than consummated. Something of a similar evolution has occurred on the
side of the independent variable. That is, early efforts were devoted
primarily to demonstrations of the consequences of deprivation of some bio-
chemical substance (e.g., oxygen) with research attention to psychosocial
deprivation on a large scale being of fairly recent origin.
Within this broad domain of seentific concern with the consequences
of deprivation, it is possible to offer some generalizations about findings
that have emerged. When the investigations have utilized an experirental
model, as is the case in animal studies where the independent variable can
be precisely quantified and manipulated, findings have tended to be fairly
consistent across a broad range of developmental influences. However, when
hypotheses are tested in humans, forcing reliance upon a field study model
with the hope of finding different levels of the independent variable in
nature, consistency disappears and polemics materialize. In field studies
the complexity of the concept of deprivation immediately becomes'apparent.
Not only is it difficult to disentangle one type of deprivation from another
(e.g., nutritional and psychosocial), but to be precise about the time of
maximum exertion of effect of one or another component is almost impossible.
That is, prenatal malnutrition tends to be correlated with postnatal mal-
nutrition and also with prenatal and postnatal psychosocial deprivation;
in field studies it is virtually impossible to determine which is associated
with any observed primary effect.'
When one uses psychosocial factors as both independent and dependent
variables in a deprivation model design, it becomes increasingly difficult
to arrive at any firm generalizations. Animal studies have not yielded
completely consistent data either within or across species, as discussed by
3
Griffin and Harlow (1966). And at the human level with the inevitable for-
feiture of the experimental method, data have been both scanty and conflicting.
Some of the variance may relate to the difficulty involved in precisely
defining what it is we are talking about when we speak of "psychosocial depriva-
tion." Although an overly pedantic attempt at definition would be both
unnecessary and tedious, it would perhaps help the discussion to follow if
some attention were given to current thinking about the concept of psycho-
social deprivation. Undoubtedly the most serious attempt at definition has
been made by Richardson (1966), who discusses the slipperiness of the con-
cept as follows;
"Two kinds of evidence can lead us to suspect deprivation. The first is
whether the child is able to perform at a given age within the level of
expectations and demands that are common to his tribe, society, or national
group.
"The second kind of evidence needed to judge a case of deprivation in-
volves the child's upbringing and experiences. . . If a child does not
receive the elements of upbringing or experience essential for development,
this is evidence of deprivation.
"These two kinds of evidence--of whether a child fails to live up to
expectations and of whether he has not had the experiences necessary to
prepare him to meet these expectations--are both needed to make a judgement
(1966, pp. 55-56). "
Richardson is describing the kind of inference that is always required
whenever information about a particular variable is sought after the fact- -
in this case, after the deprivation has presumably occurred. However, such
a definition would not be necessary for a contemporary assessment of
4
deprivation with follow -up sustained long enough to permit the effects of the
independently defined deprivation to appear. Unfortunately investigations of
the consequences of psychosocial deprivation are often begun retrospectively.
And yet contemporaneous assessments of children in potentially depriving en-
vironments should be a major research strategy. If one adhered rigidly to
the Richardson concept of deprivation such investigations would be precluded.
In addition to the tendency to become alert to deprivation only after
its presumed consequences have begun to appear, there is undoubtedly another
reason for the tendency to want to support presumed presence of deprivation
with proof of effect. Conceptually we are always troubled by false negatives,
in this case by the appearance of some children who appear to develop reasonably
well in an environment judged to be depriving in terms of a crucial component- -
children who grow big and strong on an inadequate diet, who grow up bright and
alert after a history of having been blue and almost lifeless for five minutes
following birth, who develop high achievement motivation in a sociocultural
context characterized by apathy and defeat. An immediate and possibly valid
explanation likely to be offered is that the crucial variable may not hit all
children with the same impact; i.e., some children in a family get more food
than others, or perhaps the anoxic baby had oxygen reserves that could not be
measured, or perhaps the last child had a special relationship with a power-
ful non-family member. Thus it could be asserted that it is safer to infer
deprivation only when one sees evidence that it has hit its target. This
author regards such concern for cushioning the concept as unnecessary. During
the middle ages it is doubtful that the occurrence of a plague was questioned
just because half the population managed to live. The fact that deprivation
defined in terms of a complex set of elements of experience needed for
5
development exists but occasionally misses its target does not require
proof of effect for presence of influence to be inferred: Such a require-
ment runs the risk of purporting to study the effects of an independent
variable (deprivation) on a dependent variable (development) by getting
information about the magnitude of the independent variable from the
measure of the dependent variable. Developmental misses are informative
about the range of human adaptability; they do not require any more con-
ceptual elasticizing for psychosocial factors than for any other type of
presumed developmental influence. As we learn to identify more of "the
elements of . . . experience essential for development," to determine
whether they are critical only in the presence or absence of other variables
or only if the phenomenological field and the objective field are isomorphic,
it will become easier to refine the definition of deprivation and thus
engage in anterospective studies of development in which psychosocial
deprivation might occur.
Evolution of Present Knowledge in the Field
With respect to the effects of psychosocial deprivation on the human
infant, there has unfortunately been more speculatim than investigation, or
perhaps less speculation than unwarranted inference. Research with this age
group is extremely difficult to conduct, not only because of the above
kr) described tangles inherent in the deprivation concept but also because of
a shortage of assessment techniques suited to the young child and of quali-
fied persons to apply them. Nor has there ever been a readily available
series of large numbers of potential research subjects--at least not readily
available to persons interested in this area. The well-child clinic has
6
been the only major resource, and 1117 and large it has not been utilized
fo: such studies. And, as now constituted in most communities, it would
produce a biased sample.
Much of the early work done in the area owes its impetus to the
reports of Spitz (1945, 1946), who launched what might be called the
maternal deprivation' decade. Pis report that infants who were abruptly
separated from their mothers often went into a state of depression and
showed a sharp decline in cognitive functioning seemed to catalyze Tiorld-
wide interest in the subject. Implicit in this work was the importance of
the emotional relationship existing between mother and infant. The publi-
cation of Borlby's monocraph (1952) summarizing. the existing world literature
on the subject, while still concentrating on the maternal component of the
deprivation, made concern with the psychosocial environment during infancy
official and instigated very salutary international changes in institutional
practices. Historians can allays recognize precursors to an idea T,*hose time
has come, and Stone (1954) pointed out that curiosity about the consequences
of what is here being called psychosocial deprivation is very old indeed.
In fact, just prior to the time Spitz published his first report, Dibble
(1943) had been stressing the importance of the mother for the healthy
development of the infant. however, she appeared to be placing greater
emphasis on the physical stimulation provided by the monther and thereby
missed her chance to parent a trend.
It did not take long until disclaimers appeared, occasionally caustic
and polemic (e.g., Pinneau, 1951, 1955), but more often simply as suggestions
that proponents of the maternal deprivation concept might be overstating
things in their insinuation that negative consequences were inevitable and
7
irreversible. As the exciting animal work of the mid-fifties and early
sixties came to be known, perhaps especially the work of Harlow and his
associates (1958, 1961), the adjective in "maternal deprivation" began to
be downgraded and more attention paid to the noun. Casler, for example,
published a monograph (Casler, 1961) in which he almost superciliously
ridiculed the notion that the "maternal' part of the concept was important
for anything other than the sensory stimulation supplied thereby.
The coup de grace to the indiscriminate espousal of the concept of
maternal deprivation was supplied not by polemics but by careful analyses
of empirical studies (see especially Yarrow, 1961, 1964; and Ainsworth, 1962)
which challenged the notions of inevitability and irreversibility of effect
and which pointed out some of the common design flaws inherent in most of
the available studies. From these analyses, from other empirical work
appearing at the time, and from new social concerns which suddenly galvanized
in the early sixties, the concept of maternal deprivation as an independent
variable was broadened to take in the entire psycho-social-cultural domain.
Again, as is customarily the pattern in the early stages of evolution of an
idea, there was a tendency to indict an entire package rather than look
immediately for those aspects which were -,nst Thus Riessman's book
(Riessman, 1962) launched a new term -cultural deprivation--and the types
of studies which followed seemed almost unconcerned with anything as specific
and personal as a young child's relationship with his mother. The suspected
irritants were much more social and cosmic! At the same times developmental
concerns (the dependent variable side of the paradigm) seemed to shift from
socioemotional to cognitive. The publication of Intelligence and Experience
8
by Hunt (1961), organizing a wealth of evidence relating to the effects
of experience on intelligence, and, by implication, of deprivation of
crucial components of experience, seemed to help shift interest from any
narrow selection of influences to an examination of a full array of
experiential parameters.
In this brief history here traced, it was interesting to note how
quickly the infant got lost. Not the importance of infancy, by any means;
just the infant. In fact, he seemed to disappear between Spitz and Bowlby!
For while Spitz's work had dealt with infants whom he had observed both
prior to and after separation from the mo':her, most of the studies found
by Bowlby and referred to in his volume were retrospective and involved
later reports of persons who had been separated from their mothers during
infancy. Bowlby, Ainsworth, and Rosenbluth (1956) later carried out a
follow-up study using children who had been separated from their mothers
during infancy, but the assessments were carried out during the middle
childhood period. And yet most of the available literature on maternal
deprivation seemed to suggest that the critical period for producing
maximum separation effects was during infancy.
In another vein, studies which began to appear in the early sixties
contrasting the developmental picture of children who were culturally
deprived (to use the term then fashionable) and non-deprived groups tended
to show that by the time the children reached public school age they already
functioned at a lower level than their non-deprived peers. This could only
be interpreted as indicating that the depressing effect of the environment
had already been at work during the early years of life, even though the
research groups had not studied their subjects during that period.
9
Field studies of early deprivation
Some of the early leads regarding psychosocial deprivation during
infancy came from studies concerned with effects of different child-rearing
practices, with information occasionally offering misleads rather than
heuristic clues. an early study of child-rearing practices in different
ethnic and social ,,ass groups, Davis and Havighurst (1946) suggested that
the early family environment of the lower class child was perhaps more
comfortable, in terms of parental practices, than that of the middle class
child. Lower class parents reported themselves, in interviews, to be more
indulgent in terms of early feeding practices and to be less concerned
with early toilet training than did middle class parents. Also they were
interpreted as being more willing to grant independence as their children
grew older. Class differences were more striking than race differences.
Williams and Scott (1953), using only Negro families, found essentially
the same pattern of class differences.
Less than a decade later, however, data from the study of Maccoby and
Gibbs (1954) suggested just the reverse pattern. That is, maternal inter-
views with parents of five- year --olds brought forth the information that,
during infancy as well as at the time of the interviews, the middle class
parents were more lenient and permissive than the lower class parents.
A number of independent researches subsequently carried out tended, in
general, to support the findings of the Maccoby and Gibbs study or else
to find no appreciable differences between social class groups (see.
Caldwell, 1964). Bronfenbrenner (1958) suggested that the tenor of the
postwar period was in favor of permissiveness and that middle class parents,
L_ -
10
with their greater susceptibility to opinions of the "experts," had
probably assimilated the recommendations more rapidly than lower class
parents. At any rate, certainly no data have emerged since the early
fifties to suggest that in any significant way the child from what might
be considered deprived social circumstances has any particular psycho-
logical advantage in his early family environment.
Most of these studies used as contrast groups samples that perhaps
did not differ very drastically along a hypothetical continuum from
deprived to nondeprived. For example, in the Maccoby and Gibbs (1954)
study, the groups might be described as upper-lower versus lower-middle.
In more recent years, however, careful studies of groups that could be
characterized as truly deprived have been carried out, most notable of
which are perhaps Wortis et al. (1963), Pavenstedt (1965), and Malone
(1966). In Wortis et al. study, observations were carried out during
infancy, making this perhaps the most relevant study for the present
discussion. They found that what might have been called permissiveness
in another decade appeared to be more a lack of concern. Weaning and
toilet training, rather than being carried out slowly and in line with
the child's indications of maturity, were handled inconsistently and in
the manner that caused the least trouble for the mother. These mothers
were extremely intolerant of any experssions of aggression from the child-
ren but were relatively unconcerned about manners, noise, and cleanliness
in general. Also the mothers themselves were depressed and withdrawn and
in general pessimistic about life. Independently assayed, i.e., not
judged on the basis of development of the children, psychosocial depriva-
tion was a demonstrable reality. Here it should be mentioned parenthetically
11
that all the children in this study had been born prematurely, thus
compounding biological insult with psychosocial deprivation. The children
involved in this study, now in the middle childhood years, are currently
being followed to learn about their school achievement and general social
and emotional adjustment (Wortis, personal communication).
Caldwell and Richmond (1967) have reported on a small sample of low-
income mothers whose maternal behavior was observed and rated sequentially
and whose children were examined at the same points in time. These data
offer leads regarding the relation between certain discrete maternal be-
haviors that would be considered depriving and the development of the
children during the infancy period. The results showed that, on a number
of scales purporting to measure some aspect of affiliative (emotionally
supportive) maternal behavior, mothers who were rated lowest on these
scales had children with lower IQ's at 12, 18, and 24 months of age.
Similar associations were found for a series of scales measuring different
facets of achievement motivation in the mothers. In general the correla-
tions between affiliative and achievement oriented maternal behavior and
development at 12 months of age tended to be low and positive, with both
the number of coefficients attaining statistical significance and their
magnitude increasing as the children approached two years of age.
With the intensified interest in cognitive development in the sixties
has come a renewed interest in instrument development for the early infancy
period. This interest, coupled with concern for psychosocially deprived
children, has led to a number of new studies examining early differences
between infants being reared in suspect environments and those being reared
12
in environments labeled by fiat as adequate (i.e., middle class homes).
Two recent studies have produced somewhat contradictory results. Wachs
(1967) found social class differences as early as 12 months on an ordinal
scale developed by Uzgiris and Hunt (1964) which measures attainment of
the concept object permanence. On the other hand, Golden and Birns (1968)
were unable to demonstrate social class differences in test scores of
Negro infants during the first two years of life on either a standard test
of infant development (the Cattell) or an experimental procedure developed
to measure more subtle aspects of cognitive functioning. They did, however,
report that the children from lower social class groups were more difficult
to test, with one-fourth of the children from welfare families requiring
more than one session to complete the testing. This finding, whether due
to motivational factors or to difficulty in adjusting to the interpersonal
components of the test, somewhat weakened the authors' conclusions of no
social class differences in performance.
A most interesting and important study by Decarie (1965), although
carried out in quite a different context, has yielded important data re-
lating to the effects of psychosocial deprivation during infancy. She was
concerned with determining whether the Piagetian concept of "object per-
manence" bore a relation to the psychoanalytic concept of "object relations,"
developing an experimental procedure to measure the former and adapting
certain items from the Griffiths Scale of Mental Development (1953) to
measure the latter. For her population' she used 30 home-reared infants,
30 in adoptive homes, and 30 residing in an institution. Her results showed
clearly the damaging effects of the psychosocial deprivation inherent in the
institutional atmosphere, as on both the measures of object permanence and
13
object relations the institutional infants (examined repeatedly up to 2'1
months of age) lagged behind the other two groups, showed deviant as
well as developmental profiles, and were far more variable than the
other two groups. Significantly, the children in adoptive homes, all of
whom had undergone at least one separation and relocation, generally
occupied an intermediate position between the institution and home
reared infant.
The important question of what happens developmentally during the
period between roughly the end of the first year and age three, at which
time at least a few children come under research scrutiny y remains -a
research problem of top priority. Early studies (again often done out-
side the deprivation model but offering data relevant to inferences in
this area) tended to show no major, differences between infants presumed
to be non-deprived and a group that might be considered deprived of one
or more developmental supports. For example, Knobloch and Pasamanick
(1953) showed that Negro infants were slightly accelerated in motor
development during the first year of life and certainly showed no major
deficit in functioning during this period. Yet data coming from various
compensatory education programs (see Hodges and Spicker, 1967)
dealing with deprived and non-deprived preschool children, have consis-
tently shown a deficit in functioning levels in the deprived groups
during the preschool or kindergarten period. Deutsch has suggested (1965)
that, rather than diminishing when the children reach the age of formal
14
public education, the discrepancy increases with age.
All the available studies are consistent in their implications.
During the first year of life infants from deprived and non-deprived
homes appear to develop at about the same rate; we tend to lose track
of them from roughly one to three years of age; when found a;ain :1,..
age three the deficit is striking (and depressingly resistant to sus-
tained change). Such findings suggest that the optimal time for trying
to reverse the deprivation is during the two year hiatus when subjects
are seldom visible to research scrutiny. The programs described by
Caldwell and Richardson (1963) Robinson (1968) and Gordon (1967) have
been designed to try to develop intervention strategies a9rronrinte
for this crucial period.
Undoubtedly the most significant recent addition to information in
this area is that provided by Bayley (1965). With carefully trained
examiners giving the revised Bayley Scales of Mental and Motor Development
to 1,409 infants from a representative sample of the total American popu-
lation, she has established performance curves for relevant subgroups of
infants between one and fifteen months of age. There were no significant
differences in the mental scales as a function of social class, sex, race,
or parental education up to 15 months of age. However, on the motor scales
Negro infants tended to score higher than whites, with the differences
significant at most evaluation points up to 12 months of age but not there-
after. The drop that occurs for the children from lower social class
gadulthsiaissar ,alilail.....6...44,.. ,a,e4,.. al, t ,.. ,,, - ,^, le , .......4.- ft- - ..-..,,,, xyn-re-dvil
15
backgrounds thus can be pinpointed as occurring somewhere between 15
months and three years of age. Similar data had been reported earlier
by Hindley (1960) for a small group of British children within the age
range of six to eighteen months.
Experimental studies relevant to the deprivation model
With human subjects, one cannot point to a truly experimental study
that involves deprivation. The only pattern ethically open to the inves-
tigator is to try to reverse deprivationi.e., to enrich-- and examine
the effects. The earliest important study of this nature was that of
Skeels and Dye (1939), the project that became the "sleeper" of its time.
Briefly, these authors transferred 13 young children who were showing
retarded development in an orphanage to an institution for the mentally
retarded. In this new environment, hardly one likely to be thought of as
enriching, the infants were cared for by adolescent and young adult men-
tally retarded girls. Instead of being part of the crowd of infants
having to share the scanty amount of adult attention available in the
orphanage, the children were suddenly cynosures in a population probably
starved for small, dependent creatures in need of love and attention.
The story has the happiest of endings. Most of the children, after re-
ceiving even this distorted brand of enrichment, soon became adoptable
and, when found and studied some 30 years later (Skeels, 1966) were found
to have been able to maintain themselves in the community and to have pro-
duced offspring that functioned within the normal range. At the time
16
of this writing, this is the only enrichment study known to this author
which has had such an extended follow-up.
Early on the heels of the first speculations about maternal depri-
vation (at a time when no one would have said anything good about the
Skeels and Dye work), Rheingold (1956) carried out an important study
aimed at determining whether the amelioration, if not the removal, of
psychosocial deprivation would produce favoraigt-developmental consequences.
She served as a sort of congregate mother for two sequential groups of
four infants residing in an institution and selected eight matched subjects
as controls. After two months of her special mothering of them, the in-
fants were examined on certain eye-hand tests selected from the Cattrell
Infant Test and on certain aspects of social development. The ex-
perimentally mothered infants were found on retest to be more socially
responsive, not only to their special caretaker but to other people:: on
the ward as well, and to be slightly though nonsignificantly advanced
in postural and motor behavior.
Rheingold and Bayley (1959) re-examined these infants when they
were approximately two years old. At this time they could detect no
differences between the two groups of babies, either in terms of social
behavior or performance on a developmental examination. The authors con-
cluded that, while the early enrichment was enough to produce concurrent
changes, it could not sustain them for a proldnged period. To this author,
however, the investigators seemed too conservative in their conclusion,
17
as the enriched babies were found on the follow-up to vocalize more than
the control babies. The relationship between early vocalizations and
later language is far from understood, but in view of the unyielding con-
sistency with which deprived children are shown to function less adequately
in the language area; any type of social experience which is associated
with an increase in vocalization is worthy of further study.
There are a few additional experimental studies which might be cited
as relevant, but most of them have involved what this author would con-
sider "artificial" enrichment or else have involved such brief follow-up
as to make inferences based on them much too hazardous. For example,
Casler (1965) has investigated the effects of kinesthetic stimulation
on institutional infants prior to adoption; Dennis and Sayegh (1965)
tried accelerating fine motor performance with only a two-week experimental
period; Ourth and Brown (1961) had mothers provide extra stimulation for
their infants during the neonatal period and compared frequency of crying
in this group and a routine ward control group; White and Held (1966)
reported the consequences for institutional infants of extra handling,
motility, and an enriched visual surround. While all such studies are
important, they do not give us the kind of hard data needed to understand
the effects of "real life" deprivation, or the removal of same, on sus-
tained development. The current enrichment projects involvhg infant sub-
jects (Gordon, 1967 Caldwell and Richmond, 1968; Painter, 1968; Schaefer,
1968) will in time provide some much needed information about the long-
18
term effects of enrichment upon the cognitive and socio-emotional
development of infants.
Areas of Needed Research
As the literature just cited contains so few studies concerned with
the effects of psychosocial deprivation on the human infant that have
(a) been prospective, (b) used respectably sized samples, and (c) followed
the subjects long enough to permit either cumulative deficit or natural
reparative processes to occur, it is obvious that the need for more re-
search dealing with any facet of the problem is acute. Even so, there
are priorities which could be established which should lead to greater
and perhaps more immediate returns. A few tasks which should receive
massive research efforts will be discussed here.
1. Improved techniques of assessing the psychosocial environment.
Behavioral scientists have tended to concentrate on measuring and quan-
tifying behavioral outputs, while paying little attention to the en-
vironments in which the behavior occurred--all the while passionately
espousing convictions about the importance of environmental influences for
understanding the developmental phenomena being measured. This relative
imbalance needs to be corrected by research concerned with identification
of the elements of experience which can influence development and to
studies of their incidence and distribution in populations known to con-
tain all levels of the dependent variable under scrutiny. Such a recom-
mendation does not contradict the point made earlier about try!ng to avoid
19
possible contamination of the measure of deprivation with knowledge about
the dependent variable. Rather, in the first stage of exploration of
the variables that are presumed to co-vary, there is a need for diversity
of both in order to make certain that there will be enough range to permit
a relationship to appear should one exist.
The author and her colleagues have devoted considerable time to
the development of a procedure for assessing the specifics of the dail7
home environment of infants. On the basis of the literature dealing with
patterns of family functioning in homes offering contrasting arrays of
resources (as summarized by Chilman, 1966), an inventory was developed
to provide an index of the level of stimulation and support offered a
child in his home. The inventory is scored on the basis, of a home visit
during which careful attention is paid to the specific ways that the mother
handles her child during the observation. The physical aspects of the
child's environment are also noted, and information is secured about the
breadth of home experiences available to the child, A majority of the
items relate to the specific transactions that occur between parent and
child during the visit. Such a procedure is undoubtedly insensitive in
comparison to detailed anthropological studies of the home environment but
is probably better than a compilation of demographic characteristics on
the basis of which a family is or is not designated as deprived. Such a
gross classification is perhaps tantamount tQ dichotomizing r -sons as
"normal" or "retarded" on the basis of the criterion of school achievement
20
For some purposes these gross classifications ,_re adequatee.g., knowing
how many families need vocational training programs or better housing,
or knowing how many children need remedial educational worktqt
when one is seeking to establish relationships between these two variables,
the likelihood of obtaining significant associations is greatly enhanced
by refining both classes of variables.
The procedure (Caldwell, Heider, and Kaplan, 1966) developed for
assessing the qualities of each home has been labeled the Inventory of
Home Stimulation. Early standardization data have demons:rated a wide
range of variability on the Inventory in families in whici variation can
be expected in certain key developmental variables in the children. While
lower class and middle class families obtain significantly different mean
scores, as would be predicted, a much greater range of scores wmfound
for lower class than for middle class families. This confirms the sus-
picion of many people that in some families, ostensibly quite deprived in
tangible and material assets, child care patterns are such that the impact
of the material deprivation may well be minimized. The author would
hypothesize that it is children from such families who attenuate apparent
relationships between deprivation and mental development (or educability,
achievement motivation, etc.): such families can easily be identified
by a more sensitive measure of the developmental enviromr-ato
2. Change-sensitive measures of early child devc1cpment. Another
important need, if the concept of deprivation is to be strengthened as a
21
useful scientific construct, is more attention to the development of
change-sensitive measures of human behavior. The legitimate concern
for stability in measuring instruments (e.g., test-retest reliability,
etc.) has led to a minimization of effort to develop significant mea-
sures which are exquisitely sensitive to a change in the environmental
situation. There are those who would argue that one need look no further
than to the existing developmental tests, which are perhaps more prone
to fluctuate than most test constructors would desire! Yet instruments
that rely on ratio estimates of developmental level do not qualify for
a truly change-sensitive measure. Preferably here would be naturalistic
observation techniques which perhaps involve nothing more quantitative
than a frequency count. The work of those who use operant procedures
to induce behavioral change (see Bijou and Baer, 1961) exemplifies
maximum utilization of change-sensitive behavioral measures. In this
work the experimental paradigm involves the deliberate selection of a
variable which is expected to change when maintaining or reinforcing
condition change. The procedure employed by Harris, Wolf, and Baer
(1963) will serve as an illustration even though it did not involve in-
fants. In a series of studies carried out with nursery school children,
they followed a standard format: (1) obtained a base frequency of the
behavior either to be maximized or extinguished (e.g., non-social play,
or crying); (2) instituted a systematic program of differential attention
to the child contingent upon his behavior (e.g., attended to child in-
22
volved in group play; ignored crying child) and obtained measures of out-
put; (3) reversed the pattern of differential attention (ignored group
play, or attended to child when he was crying) to make certain that any
obtained changes in output were not simply coincidental; and (4) again
offered differential attention as in (2) and determined change in fre-
quency of response. Frequency of emission of a large variety of behaviors
has been shown to change significantly when the environmental conditions
(usually adult attention or some other form of social reinforcement)
which shape or maintain the responses are changed. This model has been
successfully used in several studies of infant behavior--elicting and
extinguishing smiling (Brackbill, 1958), and head turning (Papousek, 1961)
The focus of much of the work in this area has not been environ-
mental deprivation per se but rather on demonstration of the dependence
of certain types of behavioral output upon specifiable environmental con-
ditions. However, only by the careful delineation of response measures
which can change is it possible to demonstrate the capacity of the environ-
ment to foster and maintain behavior change. These demonstrations
have been quite persuasive, and there is now a need to develop additional
change sensitive measures for infants other than simple frequency of res-
ponse emission.
3. Esglez.eLpati.onshi12tyclorfi.ntleen constitutional factors and:
the susceptibility to the influence of deprivation. In an earlier paper,
(Caldwell, 1964), this author discussed the need to consider the inherent
23
characteristics of the children exposed to different types of early ex-
perience when looking for evidence of environmental influence. This,
of course, is but a reiteration of the position taken by Thomas, Chess,
and Birch (1963) and others who have stressed the importance of con-
stitutional factors as buffers against the vagaries of experience.
To illustrate the validity of the point in this context, data
from a study by Schaffer (1966) will be cited. In a group of infants
ranging in age from one to 29 weeks of age at admission to a hospital
for a variety of illnesses, hospitalized at least seven weeks, and
medically adjudged to be completely recovered at time of discharge,
variability of individual response to the separation (described by
Schaffer as depriving) was evaluated as well as the group response. The
mean developmental quotient of the infants at the end of hospitalization
was approximately 85, but it rose to 95 within 18 days after the return
home and showed no further fise after three months. Schaffer interpreted
this as indicating that the quotients had returned to their predeprivation
levels. The range of individual post-hospitalization changes was from
-6 to +31 points. When these changes were related to ratings of activity
level made on the infants while in the hospital, it was found that the
infants showed the greatest rise (i.e., those who had presumably shown
the sharpest drop in the hospital) were those who had been least active
during their hospitalization. The two infants given ratings indicating
high activity level had both shown a slight decline in test score.
24
Schaffer suggests that DQ rise following alleviation of deprivation may
serve as an indicator of vulnerability to deprivation and that indicators
such as activity level during deprivation may give some indication of
potential resistance to environmental stress. Inactive infants would
certainly resemble the fragile ones described by Spitz (1945) as
marasmic. Other studies taking into consideration these constitutional
factors are very much needed.
4. Research utilization of the clinical single-case model.
Although the author does not wish to belabor the nutrition comparison
made earlier in this paper, and certainly not to imply that measurement
in the two domains bears much resemblance to one another, the task of
providing proof of effect of malnutrition and of psychosocial deprivation
is similar. In cases of suspected malnutrition, investigators have often
had to rely on evidence from the dependent variable to infer the magni-
tude of the independent variable--ios., to infer nutritional status
from growth for age--in that only after children are found to deviate
from growth norms do they come to the attention of persons who can hope-
fully offer any remediation. But, of course, discovery of height-
weight deviation does not always herald malnutrition, as the infant may
be simply manifesting a genetic potential for small stature. A careful
inquiry to clarify the etiology may confirm the malnutrition, and im-
mediate clinical action will likely be taken to rectify the inadequate
nutrition. In the logic of diagnostic procedure, however, subsequent
25
measures of height and weight which indicate movement toward growth norml
are not proof of the accuracy of the diagnosis. Equally necessary is
a check to determine whether food intake has actually changed during
the interim in the direction that had been prescribed.
In cases of suspected psychosocial deprivation observed clinically,
this same model has generally been followed, but often only part way.
For one thing, similar evidence--deviation of developmental rate for age
norms in the absence of evidence pointing to biological causation--it3
often the first indication that psychosocial stimulation might be inade-
quate to pace normal growth. (It is the first indication only in the
same way that appearance of a malnourished child at a clinic forces re-
cognition that the food supply of a population is inadequate.All sorts
of other signs are usually clamoring to be perceived, but until there is
clinical concern it is usually easy to overlook them.) In the second
place, attempts to alter the level of psychosocial input offer about
the only reasonable hope of remediation available; if the deviation is
due to genetic causes there is little that can be done.
Here the similarity between the two types of clinical decisions
usually breaks down, for it is much easier to specify the minimum daily
requirements of nutrition than it is of experience. Even so, remediation-
as-proof is still justified, provided one does not short circuit the pro-
cedure necessary for establishment of full proof of effect. There is
no question but that attempted remediation of the deficiencies of the
26
psychosocial environment should begin almost immediately, as all the
evidence available suggests that the longer the delay the less hope one
have for complete remediation. However, immediate remediation without
careful assessment of the environment in which the child is developing
deprives a given clinical case of its opportunity to contribute to the
needed foundation of evidence in this area. Only the full chain--deviant
development-evidence of deprivation. alteration of psychosocial environ-
ment-change in developmental ratelevidence of change of level of psy-
chosocial stimulation--can offer the needed proof of the effects of
the environment on the infant's development.
This single case model, essentially the same as that described
earlier and associated with the behavior modification groups has impor-
tant implications for clinical studies of the effects of psychosocial
deprivation. Such clinical appraisals followed by remediation can do
almost as much to buttress our knowledge of the consequences of psycho-
social deprivation as the needed epidemiological, normative, and experi-
mental studies.
Epilogue
This paper has attempted to highlight certain procedures which must
be followed if acceptable scientific proof of the power of psychosocial
deprivation to distort a child's development is to be obtained. As an
epilogue one brief note of caution should be added. That is, concerned
people tend to pair the word "deprivation" almost automatically with the
27
word "enrichment." The above cited studies using the experimental
method--all of which involved enriching rather than depriving--reveal
that the author tends to think in the same way. But enrichment is not
necessarily the antonym of deprivation. Manipulations of experience
may be useful only up to an as yet undetermined safety level. To try
to go beyond this level in our fervor to counteract deprivation may be
like giving vitamins to a well -fed child--his system will just disregard
the excess. However, it may possibly be like overloading the small
world of a premature baby with too much oxygen and thus forever damaging
the cells of his retina. This is not to suggest that some kind of
experiential hyperplasis will inevitably occur following psychosocial
enrichment, but rather to serve as a reminder of the possibility that it
could occur.
28
References
Ainsworth, M. Reversible and irreversible effects of maternal deprivation
on intellectual development. Child Welfare League of America, 1962,
42-62.
Bayley, N. Comparisons of mental and motor test scores for ages 1-15
months by sex, birth order, race, geographical location, and education
of parents. Child Development, 1965, 36, 379-411.
Bijou, S. W., & Baer,,D. M. -Child develonment. systematic.and
empirical theory. New York: Appleton-Century-Crofts, 1961.
Bowlby, J. Maternal care and mental health. Geneva, Switzerland:
World Health Organization, 1952.
Bowlby, J., Ainsworth, M., Boston, M., & Rosenbluth, D. The effects of
mother-child separation: a follow-up study. The British Journal of
Medical Psychology, 1956, 29, 211-247.
Brackbill, Y. Extinction of the smiling response in infants as a function
of reinforcement schedule. Child Development, 1958, 29, 115-124.
Bronfenbrenner, U. Socialization and social class through time and space.
In E.. E, :faccoby, T. M. Newcomb, and E. Hartley (Eds.), Readings in
social psychology. New York: Holt, Rinehart & Winston, 1958. Pp.400-h25.
Caldwell, B. M. The effects of infant care. In L. Hoffman and L. Nr,
Hoffman (Eds.), Review of Child Development Research, Volume I. New
York: Russell Sage Foundation, 1964, ?Pp. 4-87.
29
Caldwell, B. M., Heider, J. 1(aplan, B.. The Inventc-ry of Eme StimU-
lation. Paper presented at the meeting of the American Psychological
Association, New York, 1966.
Caldwell, B. M., & Richmond, J. B. Social class level and stimulation
potential of the home. In J. Heilmuth (Ed.), Exceptional Infant, Vol. 1,
1967, 455 466.
Caldwell, B. M., & Richmond, J. Bo The Childrens Center in Syracuse,
New York. In L. L. Dittman (Ed.), Early child care: The new perspec-
tives. New York: Atherton Press, 1968. Pp. 326358.
Casler, L. Maternal deprivation': a critical review of the literature.
Monographs of the Socieq_for Research in Child Development, 1961,
26(2), 64 pp.
Casler., L., The effects of extra tactile stimulation on a group of in-
stitutionalized infants. Genetic Psychology Monographd, 1965, 71(1),
137-175.
Chilman, C. S, Growins_Alamps. Washington, D. C.: U. S. Department of
Health, Education, and Welfare, Welfare Administration, 1966.
Davis, A., & Havighurst, R. J. Social class and color differences in
child. rearing. American Sociological Review, 1946, 11, 698-710.
Decarie, T. G. Intelligence and effectivitilp2fslys12ildhood. New
York: International Universities Press, 1965.
Dennis, Wo, & Sayegh, Y. The effect of sup-plementary experiences upon
the behavioral development of infants in institutions. Child Develop-
ment, 1965, 38, 81-900
30
Deutsch, M. The role of social class in language development and cogni-
tion. AliJournericat)rthoschiatr, 1965, 38, 78-88.
Golden, M., & Birns, B. Social class and cognitive development in infancy.
Merrill-Palmer Quarterly, 1968, 14, 139-149.
Gordon, I. J. Home stimulation for disadvantaged infants. Unpublished
manuscript, 1967.
Griffin, G. A., & Harlow, H. F. Effects of three months of total social
deprivation on social adjustment and learning in the Rhesus monkey.
Child Development, 1966, 37, 533-547.
Griffiths, R. The abilities of babies. New York: McGraw-Hill Book Coe,
1954.
Hindley, C. B. The Griffiths Scale of Infant Development: scores and
predictions from 3 to 18 months. Journal of Child Psychology and
PsychiatryL 1960, 1, 99-112.
Harlow, H. F. The nature of love. American Psychologist, 1958, 13,
673-685.
Harlow, H. F., & Harloe, M. K. A study of animal affection. Natural
History., Dec., 1961, 48-55.
Harris, F. R., Wolf, M. Me, & Baer, D. M. Effects of adult scici9.1
forcement on child behavior. In W. W. Hartup and Nancy L. Smothergill
(Eds.), The young, child: Reviews of research. Washington, D. C.,
National Association for the Education of Young Children, 1967/ pp. 13 -26.
Hodges, V. L.. & Spicker, H. H. The effects of preschool experiences on
culturally deprived children. In W. W. Hartup and Nancy L. Smithergill
(Eds.), The young child: Reviews of research. Washington, D. C.,
National Association for the Education of Young Children, 1967.
Pp. 262-289.
31
Knobloch, E., & Pasamanick, B. Further observations on the behavior
development of Negro children. Journal of Genetic Psychology, 1953,
83, 137-157.
Maccoby, E., & Gibbs, P. K. Methods of child-rearing in two social
classes. In W. E. Martin and Celia B. Stendler (Eds.), Readings in
Child Development. New York: Harcourt, Brace & Co., 1951.. Pp. 380-396.
Malone, C. A. Safety first: comments on the influence of external danger
in the lives of children of disorganized families. Arne/lean Journal
of Orthopsychiatry, 1966, 36, 3-12.
Painter, G. Infant education. San Rafael, Cal., Dimensions Publishing
Co., 1968.
Papousek, H. Conditioned head rotation reflexes in infants in the
first months of life. Acta Paediatrica, 1961, 50, 565-576.
Pavenstedt, E. A comparison of the child-rearing environment of upper=
lower and very low-lower class families. American Journal of Ortho-
psychiatry, 1965, 35, 89-98.
Pinneau, S. R. A critique on the articles by Margaret Ribble. Child
Development, 1951, 21, 203-228.
Pinneau, S. R. The infantile disordets of hospitalism and anaclitic de-
pression. Psychological Bulletin, 1955, 52, 429-452.
Rheingold, H. L. The modification of social responsiveness in institu-
tional babies. Monograph of Society for Research in Child Development,
1956, 21, Serial No. 63, No. 2.
32
Rheingold, H., & Bayley, N. The later effects of an experimental mod-2_-
fication of mothering. Child Development, 1959, 30, 363-372.
Rhinegold, H. L., Gewirtz, J. L., & Ross, H. W. Social conditioning
of vocalizations in the infant. Journal of Comparative and Physio10-
gical Psychology, 1959, 52, 68-73.
Ribble, M. A. The rights of infants. New York: Columbia University
Press, 1943.
Richardson, S. A. Psychosocial and cultural deprivation in psychobiol-
logical development: Psychosocial aspects. In Deprivation in psycho-
biological development. Pan American Health Organizations Scientific
Publication No. 134, 1966.
Ries swan, F. The culturally deprived child. New York: Harper & Row,
1962.
Robinson, H. B. The Frank Porter Graham Child Development Center. In
L. L. Dittman (Ed.), Early ralilLarglita.tEinruIsLia9a. New York:
Atherton Press, 1968. Pp. 302-312.
Schaefer, E. S., & Furfey, P. Intellectual stimulation of culturally
deprived infants. Unpublished Progress Report, 1968.
Schaffer, H. R. Activity level as a constitutional determinant of in-
fantile reaction to deprivation. Child Development, 1966, 37, 595-602.
Skeels, H. M. Adult status of children with contrasting early life
experiences. Monograph of the Society for Research in Child Developnent,
1966, 31, (No.3) pp. 1-65.
33
Skeels, H. M., & Dye, H. B. A study of the effects of differential
stimulation on mentally retarded children. Proceedings of the American
Association for Mental Deficiency, 1939, 44, 114-136,
Spitz, R. A. Hospitalism: an inquiry into the genesis of psychiatric
conditions in early childhood. In 0. Fenichel, et al. (Eds.), Psycho-
analytic Studies of the Child, Volt 1. New York: International Univer-
sities Press, 1945. Pp. 53-74.
Spitz, R. A. Hospitalism: a follow-up report. In 0. Fenichel, et al.
(Eds.), Psychoanalytic Studies of the Child, Vol. 2. New York: Inter-
national Universities Press, 1946. Pp. 113-117.
Stone, L. J. A critique of studies of infant isolation. Child Develop-
ment, 1954, 25, 9-20.
Thomas, A., et al. Behavioral individuality in early childhood. New
York: New York University Press, 1963.
Uzgiris, I. C., & Hunt, J. McV. A scale of infant psychological devel-
opment. Unpublished manuscript 1964.
Wachs, T. D. Cognitive development in infants of different age levels
and from different environmental backgrounds. Unpublished manuscript,
1967.
Weisberg, P. Social and nonsocial conditioning of infant vocalizations.
Child Development, 1963, 34, 377-388.
White, B. L., & Held, R. Plasticity of sensorimotor development in the
human infant. In I. F. Rosenblith and W. Allinsmith (Eds.), The causes
of behavior II. Boston: Allyn and Bacon, 1966. Pp. 60-70.
34
Williams, J. R., & Scott, R. B. Growth of Negro Infants: IV. Motor
development and its relationship to child-rearing practices in two
groups of Negro infants. Child Development, 1953, 24, 103-121.
Wortis, H., et al. Child -- rearing practices in a low socioeconomic
group. Pediatrics, 1963, 32, 298-307.
Yarrow, L. J. Maternal deprivation: Toward an empirical and conceptual
re-evaluation. Psychological Bulletin, 1961, 58, 459-490.
Yarrow, L. J. Separation from parents during early childhood. In
M. L. Hoffman and L. W. Hoffman (Eds.), Review of child develop-
ment research, Vol. I. New York: Russell Sage Foundation, 1964
Pp. 89-130.
35
Footnotes
* Center for Early Development and Education
814 Sherman Street, Little Rock, Arkansas 72202
1. This paper was prepared under a contract with
the National Institute of Child Health and Human
Development as part of the background work for a
task force concerned with the effects of psychosocial
deprivation.
4